Metoda inverzne kinematike s pomočjo genetskega algoritma

magistrsko delo

Matjaž Petek (Author), Riko Šafarič (Mentor), Karl Gotlih (Mentor), Suzana Uran (Co-mentor)

Abstract

Zaključno delo predstavlja dva nova algoritma za reševanje inverzne kinematike mehanizma robotske roke, ki delujeta na podlagi genetskega algoritma. Algoritma sta razvita in delujoča za splošno robotsko roko, ki je grajena iz sklepov in ročic. Glavna motivacija dela je sicer usmerjena v optimizacijo algoritmov za uporabo pri humanoidnih robotskih mehanizmih. Le-ti bi lahko z uporabo algoritmov posnemali človeško nebesedno komunikacijo, tj. kretnje rok, premikanje telesa itd. Algoritem za izračun inverznega kinematičnega modela na podlagi genetskega algoritma s poznavanjem Denevit-Hartenbergovih matrik (IKM-GA-DH) omogoča zanesljivo reševanje inverzne kinematike tudi za redundantne robotske roke, vendar pa za to potrebuje direktni kinematični model za vsak opazovani sklep. Algoritem za izračun inverznega kinematičnega modela na podlagi genetskega algoritma s poznavanjem direktnega kinematičnega modela (IKM-GA-DKM) pa za reševanje inverzne kinematike potrebuje le direktni kinematični model vrha opazovanega mehanizma, vendar pa rezultati pokažejo, da zato ni sposoben reševati redundantnih mehanizmov, ki imajo več redundantnih prostostnih stopenj, enako natančno kot jih lahko algoritem IKM-GA-DH.

Keywords

inverzna kinematika;redundantna robotska roka;genetski algoritem;magistrske naloge;

Data

Language:	Slovenian
Year of publishing:	2021
Typology:	2.09 - Master's Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[M. Petek]
UDC:	004.8.021:007.52(043.2)
COBISS:	97754883
Views:	106
Downloads:	25
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Inverse kinematics method based on genetic algorithm
Secondary abstract:	In this thesis work, two new algorithms for solving an inverse kinematic of a robotic arm mechanism are presented. Both algorithms are based on Genetic algorithm. Mentioned algorithms are developed for a general robotic arm mechanism that consists of joints and links. Main motivation of the thesis work is directed more towards optimization of algorithms for use in humanoid robots. The humanoid robots would be able to use the algorithm to repeat the unspoken language of human gestures. Algorithm for solving inverse kinematic model based on genetic algorithm with Denevit-Hartenberg matrixes (IKM-GA-DH) is capable to solve the inverse kinematic for a redundant mechanism based on complete forward kinematic model for every joint of the observed mechanism. Algorithm for solving inverse kinematic model based on genetic algorithm with forward kinematic model (IKM-GA-DKM), on the other hand, can solve the inverse kinematic based on forward kinematic model of only end-effector. But results show that algorithm IKM-GA-DKM isn’t capable of solving redundant mechanisms, with many redundant DOF with the same precision as algorithm IKM-GA-DH.
Secondary keywords:	inverse kinematics;redundant robotic arm;genetic algorithm;
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Mehatronika
Pages:	1 spletni vir (1 datoteka PDF (XI, 80 f.))
ID:	13892914